基于图卷积网络的多信息融合驾驶员分心行为检测*

doi:10.19562/j.chinasae.qcgc.2020.08.005

汽车工程 ›› 2020, Vol. 42 ›› Issue (8): 1027-1033.doi: 10.19562/j.chinasae.qcgc.2020.08.005

基于图卷积网络的多信息融合驾驶员分心行为检测^*

白中浩^1,2, 王韫宇¹, 张林伟¹

1.湖南大学,汽车车身先进设计制造国家重点实验室,长沙 410082;
2.福建工程学院,福建省汽车电子与电驱动重点实验室,福州 350118

收稿日期:2019-09-19 出版日期:2020-08-25 发布日期:2020-09-24
通讯作者: 白中浩,教授,E-mail:baizhonghao@163.com。
基金资助:
*国家自然科学基金(51621004,51475153)和福建工程学院科研创新平台开放基金(KF-X18001)资助。

Driver Distraction Behavior Detection with Multi-information Fusion Based on Graph Convolution Networks

Bai Zhonghao^1,2, Wang Yunyu¹, Zhang Linwei¹

1. Hunan University, State Key Laboratory of Advanced Design and Manufacturing for Vehicle Body, Changsha 410082;
2. Fujian University of Technology, Fujian Key Laboratory of Automotive Electronics and Electric Drive, Fuzhou 350118

Received:2019-09-19 Online:2020-08-25 Published:2020-09-24

摘要/Abstract

摘要： 为减少由驾驶员分心造成的交通事故,并检测驾驶员在自动驾驶情况下的分心状态以判断驾驶员是否有接管车辆的能力,提出了一种基于图卷积的多信息融合驾驶员分心行为检测方法。通过分析驾驶员分心行为和姿态特征,设计了驾驶员姿态估计图,基于图卷积网络对驾驶员姿态估计图进行特征提取,使用全连接层对所提取特征进行行为分类,同时融合手机等关键物体信息对驾驶员分心行为进行再判断。实验结果表明,本文提出的方法在SrateFarm数据集和自制数据集上分别达到了90%和93%的准确率,检测速度约为20帧/s,准确性和实时性均达到检测要求。

关键词: 驾驶员分心, 姿态估计, 行为识别, 图卷积网络

Abstract: In order to reduce the traffic accidents caused by driver's distraction, and detect the distraction state of the driver during automatic driving for judging whether the driver has the ability to take over the vehicle, a multi-information fusion driver distraction behavior detection method based on graph convolution is proposed. By analyzing the distraction behavior and posture features of the driver, the posture estimation graph of the driver is designed. The feature extraction of driver's posture estimation graph is conducted based on graph convolution network, the features extracted are classified by using full connected layer, and the driver's distraction behaviors are judged again by the fusion of key objects like mobile phone. Experiment results show that the method proposed can achieve 90% and 93% accuracy on SrateFarm dataset and the self-made dataset respectively with a detection speed of 20 frames per second, meeting the detection requirements of accuracy and real-time performance.

Key words: driver distraction, posture estimation, behavior recognition, graph convolution network

白中浩, 王韫宇, 张林伟. 基于图卷积网络的多信息融合驾驶员分心行为检测^*[J]. 汽车工程, 2020, 42(8): 1027-1033.

Bai Zhonghao, Wang Yunyu, Zhang Linwei. Driver Distraction Behavior Detection with Multi-information Fusion Based on Graph Convolution Networks[J]. Automotive Engineering, 2020, 42(8): 1027-1033.

参考文献

[1] DINGUS T A, KLAUER S G, NEALE V L, et al. The 100-car naturalistic driving study, phase Ⅱ- results of the 100-car field experiment[R]. Blacksburg, NHTSA,2006.
[2] 马雷,陈珂,王明露,等.车路协同环境下驾驶员行为识别方法研究[J].汽车工程,2018,40(11):1330-1338.
[3] 刘志强,吴雪刚,倪捷,等.基于HMM和SVM级联算法的驾驶意图识别[J].汽车工程,2018,40(7):113-119.
[4] DING M, SUZUKI T, OGASAWARA T. Estimation of driver's posture using pressure distribution sensors in driving simulator and on-road experiment[C].2017 IEEE International Conference on Cyborg and Bionic Systems (CBS), IEEE,2017:215-220.
[5] 陈骥驰,王宏,王翘秀,等.基于脑电信号的疲劳驾驶状态研究[J].汽车工程,2018,40(5):515-520.
[6] BRAUN A, FRANK S, MAJEWSKI M, et al.CapSeat: capacitive proximity sensing for automotive activity recognition[C]. Proceedings of the 7th International Conference on Automotive User Interfaces and Interactive Vehicular Applications, ACM,2015:225-232.
[7] DEMIRDJIAN D, VARRI C. Driver pose estimation with 3D time-of-flight sensor[C].2009 IEEE Workshop on Computational Intelligence in Vehicles and Vehicular Systems, IEEE,2009:16-22.
[8] XING Y, LV C, ZHANG Z, et al. Identification and analysis of driver postures for in-vehicle driving activities and secondary tasks recognition[J]. IEEE Transactions on Computational Social Systems,2017,5(1):95-108.
[9] SESHADRI K, JUEFEI-XU F, PAL D K, et al. Driver cell phone usage detection on strategic highway research program (shrp2) face view videos[C]. Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition Workshops,2015:35-43.
[10] State farm distracted driver detection[G/OL]. https: //www.kaggle.com/c/state-farm-distracted-driver-detection,2016.
[11] ABOUELNAGA Y, ERAQI H M, MOUSTAFA M N. Real-time distracted driver posture classification[J/OL]. https://arXiv.org/abs/1706.09498,2017.
[12] BAHETI B, GAJRE S, TALBAR S. Detection of distracted driver using convolutional neural network[C]. Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition Workshops,2018:1032-1038.
[13] OU C, OUALI C, KARRAY F. Transfer learning based strategy for improving driver distraction recognition[C]. International Conference Image Analysis and Recognition. Springer, Cham,2018:443-452.
[14] KIPF T N, WELLING M. Semi-supervised classification with graph convolutional networks[J]. arXiv Preprint arXiv:1609.02907,2016.
[15] YAN S, XIONG Y, LIN D. Spatial temporal graph convolutional networks for skeleton-based action recognition[C]. Thirty-Second AAAI Conference on Artificial Intelligence,2018.
[16] ZHANG X, XU C, TIAN X, et al. Graph edge convolutional neural networks for skeleton based action recognition[J]. arXiv Preprint arXiv:1805.06184,2018.
[17] CAO Z, SIMON T, WEI S E, et al. Realtime multi-person 2d pose estimation using part affinity fields[C]. Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition,2017:7291-7299.
[18] REDMON J, FARHADI A. Yolov3: an incremental improvement[J]. arXiv Preprint arXiv:1804.02767,2018.

基于图卷积网络的多信息融合驾驶员分心行为检测^*

Driver Distraction Behavior Detection with Multi-information Fusion Based on Graph Convolution Networks

PDF (PC)

赞

可视化

摘要/Abstract

引用本文

使用本文

参考文献

相关文章 7

Metrics

本文评价

推荐阅读 10

[1]	胡宏宇, 黎烨宸, 张争光, 曲优, 何磊, 高镇海. 基于多尺度骨架图和局部视觉上下文融合的驾驶员行为识别方法[J]. 汽车工程, 2024, 46(1): 1-8.
[2]	杨彪, 韦智文, 倪蓉蓉, 王海, 蔡英凤, 杨长春. 基于动作条件交互的高效行人过街意图预测[J]. 汽车工程, 2024, 46(1): 29-38.
[3]	吕彦直,魏超,何元浩. 基于GCN和CIL的端到端自动驾驶换道方法[J]. 汽车工程, 2023, 45(12): 2310-2317.
[4]	张斌,付俊怡,夏金祥. 基于类间距优化的分心驾驶行为识别模型训练方法[J]. 汽车工程, 2022, 44(2): 225-232.
[5]	蔡英凤,吕志军,孙晓强,王海,刘擎超,陈龙,袁朝春. 基于并线行为识别的自适应巡航控制方法[J]. 汽车工程, 2021, 43(7): 1077-1087.
[6]	蔡英凤, 邰康盛, 王海, 李祎承, 陈龙. 无人驾驶汽车周边车辆行为识别算法研究^*[J]. 汽车工程, 2020, 42(11): 1464-1472.
[7]	谢枫, 娄静涛, 赵凯, 齐尧. 基于行为识别和曲率约束的车辆轨迹预测方法研究^*[J]. 汽车工程, 2019, 41(9): 1036-1042.